This continuation proposal focuses on closing the gap between comprehension of discourse in speech and print which is characteristic of unskilled readers. The novelty of this project is to bring together the knowledge base on reading differences and advanced techniques for studying the reading process in word and sentence processing, targeting a largely neglected population of non-college bound young people. Our findings to date confirm that limitations in speed and accuracy of word decoding play a large role in reading comprehension difficulties in this group, but limitations in vocabulary also play a significant role. We hypothesize that both kinds of limitations contribute to a bottleneck that prevents working memory from being used efficiently to support integrative processes in reading sentence material. One aim is to expand the bottleneck hypothesis to accommodate the new finding that vocabulary differences capture unique variance in reading comprehension beyond that contributed by listening comprehension. New studies will examine the possibility suggested by an activation-based model that this phenomenon reflects shallow encoding and slow retrieval of lexical information. Another aim is to test the hypothesis that poor readers'syntactic difficulties also reflect retrieval problems from this source. A third aim is to test the hypothesis that difficulties of learning sequences of familiar words may underlie both word-level and sentence level deficiencies of poor readers. This project addresses the causes, and by implication, the remedies for reading failure, which is acknowledged to be a major public health problem in the US and other countries. By focusing on the sources of adult reader differences in reading comprehension this project addresses the need for an informed citizenry that can meet the challenges of participatory democracy and the challenges of the work place.